Latent transitions in models like Dreamer are biased toward dense regions, creating attractors that hide true dynamics discrepancies and cause epistemic uncertainty to be unreliable while overestimating rewards.
write newline
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
UPSi integrates uncertainty from probabilistic ensemble dynamics models into predictive safety filters by formulating outcomes as reachable sets with an explicit certainty constraint for safer model-based RL exploration.
Artifacts in the environment can reduce the memory an RL agent needs to represent its history, as shown by a mathematical proof and experiments with spatial paths.
citing papers explorer
-
Biased Dreams: Limitations to Epistemic Uncertainty Quantification in Latent Space Models
Latent transitions in models like Dreamer are biased toward dense regions, creating attractors that hide true dynamics discrepancies and cause epistemic uncertainty to be unreliable while overestimating rewards.
-
Uncertainty-Aware Predictive Safety Filters for Probabilistic Neural Network Dynamics
UPSi integrates uncertainty from probabilistic ensemble dynamics models into predictive safety filters by formulating outcomes as reachable sets with an explicit certainty constraint for safer model-based RL exploration.
-
Artifacts as Memory Beyond the Agent Boundary
Artifacts in the environment can reduce the memory an RL agent needs to represent its history, as shown by a mathematical proof and experiments with spatial paths.